Class TestColumnSizes

java/tools/src/test/org/apache/orc/tools/TestColumnSizes.java:44–98

import static org.junit.jupiter.api.Assertions.assertTrue;

public class TestColumnSizes implements TestConf {
  private Path workDir = new Path(
      Paths.get(System.getProperty("test.tmp.dir"), "orc-test-sizes").toString());
  private FileSystem fs;

  @BeforeEach
  public void openFileSystem() throws Exception {
    fs = FileSystem.getLocal(conf);
    fs.mkdirs(workDir);
    fs.deleteOnExit(workDir);
  }

  @Test
  public void testSizes() throws Exception {
    TypeDescription schema = TypeDescription.fromString("struct<x:int,y:string>");
    Map<String, Integer> fileToRowCountMap = new LinkedHashMap<>();
    fileToRowCountMap.put(workDir + File.separator + "test-sizes-1.orc", 10000);
    fileToRowCountMap.put(workDir + File.separator + "test-sizes-2.orc", 20000);
    for (Map.Entry<String, Integer> fileToRowCount : fileToRowCountMap.entrySet()) {
      Writer writer = OrcFile.createWriter(new Path(fileToRowCount.getKey()),
          OrcFile.writerOptions(conf)
              .setSchema(schema));
      VectorizedRowBatch batch = schema.createRowBatch();
      LongColumnVector x = (LongColumnVector) batch.cols[0];
      BytesColumnVector y = (BytesColumnVector) batch.cols[1];
      for (int r = 0; r < fileToRowCount.getValue(); ++r) {
        int row = batch.size++;
        x.vector[row] = r;
        byte[] buffer = ("byte-" + r).getBytes();
        y.setRef(row, buffer, 0, buffer.length);
        if (batch.size == batch.getMaxSize()) {
          writer.addRowBatch(batch);
          batch.reset();
        }
      }
      if (batch.size != 0) {
        writer.addRowBatch(batch);
      }
      writer.close();
    }

    PrintStream origOut = System.out;
    ByteArrayOutputStream myOut = new ByteArrayOutputStream();
    // replace stdout and run command
    System.setOut(new PrintStream(myOut, false, StandardCharsets.UTF_8));
    ColumnSizes.main(conf, new String[]{"--summary", workDir.toString()});
    System.out.flush();
    System.setOut(origOut);
    String output = myOut.toString(StandardCharsets.UTF_8);
    assertTrue(output.contains("Total Files: 2"));
    assertTrue(output.contains("Total Rows: 30000"));
    assertTrue(output.contains("Percent  Bytes/Row  Name"));
    assertTrue(output.contains("_stripe_footer"));
  }
}